While cross-sectional studies show that drug abuse, impulsivity, serotonin dysregulation, and stressful life events are associated with suicide, data regarding the specific relationships among these factors are limited. A recent NIDA-sponsored workgroup identified gaps in this literature, set priorities for future research, and called for studies examining: (a) the role of impulsivity in suicidal behaviors and drug abuse; (b) how biological mechanisms (e.g., gene-environment interactions) predict these relationships; and (c) how suicidal behaviors and drug abuse interrelate longitudinally. Studies have typically related individual measures of impulsivity or serotonin (5-HT) markers to self-reports of previous suicidal behaviors and drug abuse cross-sectionally, but they have not determined the predictive validity of these measures in determining future drug abuse and suicidal behaviors. Furthermore, the relationships between impulsive behavior and 5-HT, while often included in theories of suicide, have not been firmly established. Additionally, it is not known how the combination of factors (e.g., impulsivity, 5-HT, and stressful life events) defined within these models affects the developmental trajectories of drug abuse and suicidal behaviors. The purpose of this longitudinal study is to examine the interrelationships among impulsivity, 5-HT, stressful life events and the outcomes of drug use and suicidality in high-risk adolescents. Our goal is to determine the direct and interactive contributions of these factors to the developmental trajectories of suicidal behavior and drug abuse. The four specific aims are to: (1) determine how self-reported histories of suicidal and drug use behaviors prior to hospitalization relate to behavioral impulsivity, stressful life events, and 5-HT markers measured at baseline; (2) establish the predictive validity of behavioral impulsivity, stressful life events, and 5-HT markers (both individually and their interaction) for determining future suicidal and drug use behaviors; (3) identify distinct types and trajectories of suicidal and drug use behaviors (using latent class modeling), and determine how groups defined by these trajectories differ in behavioral impulsivity, stressful life events, and 5-HT function; and (4) determine the relative contributions of impulsivity, stressful life events, 5-HT function, and drug use to suicidal behaviors using the theoretical framework of a leading model of suicide, and to determine the directionality of these relationships while adjusting for other known risk factors. Relevance to Public Health: Knowledge gained from this prospective longitudinal study will provide insights into the state and trait characteristics, and the 5-HT markers of adolescents with and without suicidal behaviors and substance abuse, that will enable us to distinguish developmental subgroups within this heterogeneous, high-risk population. In turn, this knowledge will be used to inform future development of treatment and prevention programs for youths with suicidal and drug use behaviors and will allow for multiple treatment approaches to be designed that take into account the specific needs of the various developmental subgroups.